A Scalable Event Dispatching Library for Linux Network Servers
|
|
- Charlotte Hancock
- 5 years ago
- Views:
Transcription
1 A Scalable Event Dispatching Library for Linux Network Servers Hao-Ran Liu and Tien-Fu Chen Dept. of CSIE National Chung Cheng University
2 Traditional server: Multiple Process (MP) server A dedicated process to every connection. Concurrency is provided by OS. Disadvantage Context-switching overhead Synchronization overhead TLB miss rate Example Apache Process 1 Process 2 Kernel Connection 1 Connection 2
3 Modern server: Single Process Event Driven (SPED) server User level concurrency within a single process. Nonblocking I/O A mechanism to detect I/O events Disadvantage Influenced by inefficient design of event dispatching mechanism in OS kernel A single page fault or disk read suspends whole server. Example Squid web cache
4 Event-driven server architecture User thread context or Connection processing context Single process Kernel Connections
5 Problem statement and our goal The problem Inefficiency design of event dispatching mechanism Our goal Improve the performance of SPED servers in user-mode.
6 A list of event dispatching mechanisms Scalable Devpoll (Solaris 8) RT signals (Linux) I/O completion port (Windows 2000) Non-scalable select() (POSIX) poll() (POSIX)
7 poll() and select() Event dispatching mechanisms in Linux select() or poll() scales poorly with large set of file descriptors. 30% of CPU time are spent on select() in a overloaded proxy server (Banga 99)
8 Interface of select() and poll()
9 Source of overhead on poll() Step by step: Read/write handler Poll() is called, array copy Linear scan in kernel Driver poll callback Poll() return, array copy Linear scan in app. Read or write handler ev_array ufds User Kernel Device driver poll()
10 Source of overhead on poll() Linear scan of ev_array and call device driver s poll callback function. Linear scan in server application to detect network events. Most work are wasted for idle connections.
11 POSIX.4 Real Time Signal Event dispatching mechanisms in Linux POSIX.4 RT extension allows signal delivery with a payload. sigwaitinfo(), sigtimedwait() Payload can carry sender PID, UID, etc. Linux extension of POSIX.4 RT signal Allow delivery of socket readiness via a particular real time signal. Pro Official support in Linux kernel since version 2.4 Con Linux specific
12 Flowchart of POSIX.4 Real Time Signal Network events Sig no FD no event type RT signal queue kernel user sigtimedwait() Read/write a FD
13 Problems in RT signals Edge-triggered readiness notification Signal queue may contain multiple events of a FD Stale event Kernel signal queue size limit entries RT signal queue overflow Some connections will fall into deadlock state.
14 Solution to RT signal queue overflow Application solution Server falls back to traditional select() or poll() Kernel solution Signal-per-fd [Chandra 01] Collapse events of the same file descriptor (signal queue size == max no. of files a process can open) => no signal queue overflow
15 Our solution to reduce poll() overhead: Scalable event dispatching library Motivation Web connections are idle most of the time. Network events in a single HTTP transaction are bursty. Our strategy Save the time wasted on idle connections for more useful works. The frequency of calling poll() on a idle connection is decreased. Average number of file descriptors at every poll() is decreased. Exploit temporal locality among events in a connection to reduce the frequency of calling poll() on a FD.
16 Live counter and polling sets: Keys to implement locality poll() A live counter is associated with a file descriptor. On every poll() to the FD: If event is detected, counter increases. If no event is detected, counter decreases. All FDs in a server are divided into three sets according to their live counter. The frequency of calling poll() on each set is different.
17 Architecture view of event dispatching library (FDM) Modular Hashing key = fd, value=nth set fdm_reg_fd(int fd) fdm_update_fd_mode(struct pollfd *fds, void *payload) fdm_wait_event() Active polling set Doze polling set Idle polling set Actual polling set sys_poll() User space Kernel space
18 Library interface design logic Reduce copy of FD array Interesting set are built gradually Separate routine to fetch event Reduce linear scan of FD array in server code Timeout timestamp and payload associated with a file descriptor is maintained in this library. Listening socket descriptor can be locked in the active polling set.
19 Proposed library interface
20 Performance evaluation Compare event dispatching mechanisms available in Linux 2.4, including our FDM library Test server dphttpd + event-threading + FDM All tests are under Multi-accept implementation Pentium III 600MHZ, 128MB RAM Test Client Httperf, HTTP workload generator three Pentium III 1GHZ, 512MB RAM
21 Dispatching overhead Goal see the overhead under Large number of idle connections Request rate is fixed at a pretty light load.
22 Server reply rate with fixed light load Reply Rate (1000 req/s, N=3, 1KB reply size) 1040 poll FDM RT signals 1020 Reply rate (reply/s) Number of idle connections
23 Server response time with fixed light load Response Time (1000 req/s, N=3, 1KB reply size) poll FDM RT signals 14 Response time (ms) FDM is more scalable than poll Number of idle connections FDM polling set maintain overhead >= polling time saved RT signal still needs to maintain timeout array
24 Dispatching Throughput Goal See the throughput under Fixed number of idle connections Overloaded request rate
25 Server reply rate with 1000 idle connections Reply rate (reply/sec) Reply Rate (1000 idle connections, N=3, 1KB reply size) poll FDM RT signals(poll) Request rate (req/sec) 100Mb Ethernet saturated, poll performs well because of multi-accept
26 Server response time with 1000 idle connections Response time (ms) Response Time (1000 idle connections, N=3, 1KB reply size) 10 poll 9 FDM RT signals(poll) Request rate (req/sec) Even at light load, we can observe overhead of 1000 idle connection. FDM suffers too since it depends on poll() Difference Between FDM And poll
27 Server reply rate with 6000 idle connections 8000 Reply Rate (6000 idle connections, N=3, 1KB reply size) Reply rate (reply/sec) RTsig+select 5000 is 4000 better than RTsig+poll, 3000 but signal queue 2000 overflow poll 1000 FDM still limit RT signals(poll) RT signals(select) scalability Request rate (req/sec) RT signal queue start overflow
28 Server response time with 6000 idle connections Response time (ms) Response Time (6000 idle connections, N=3, 1KB reply size) 200 poll 180 FDM RT signals(poll) RT signals(select) Request rate (req/sec) Signal queue overflow, server fall back to poll(), poll overhead at 6000 idle connections is a lot larger such that this behavior can t be observed at 1000 idle connections.
29 Summary of performance evaluation At light load FDM incurs low overhead At heavy load FDM improves poll() RT signal queue overflow should be protected by select(), not poll()
30 Conclusion FDM library improves poll() exploiting temporal locality property of events in a file descriptor. Better performance than RT signals if RT signal queue overflowed. RT signal queue overflow recover select() is a better choice. Availability of FDM and test web server
31 Backup slides follow
32 Keys to scalable event dispatching mechanisms Interesting set of file descriptors is built gradually inside kernel Separate building of interesting set from event retrieval Only return a file descriptor if there is a event. Collect event efficiently Cache device driver poll result Don t run driver s poll() at every poll() Or, maintain a event queue and collect events gradually at every call to TCP/IP event handler.
33 Summary of event dispatching mechanisms methods features Scalable to large set of file descriptors Event collapsing Dequeue multiple event per system call Event queue overflow When queue overflow, kernel fallback to traditional poll() Return initial state when declare interest fd Multiple interest sets maintained in kernel for per process Select() No NA Yes NA NA NA NA Poll() No NA Yes NA NA NA NA /dev/poll Yes NA Yes NA NA NA Yes RT signals Yes No No Yes No No Yes RT sig-per-fd Yes Yes No No NA No Yes Declare_interest Yes Yes Yes Yes Yes Yes No
34 Polling frequency of a set Depends on the default value of live counter Assume default value is N On every fdm_wait_event() Active polling set is polled On every N fdm_wait_event() Doze polling set is polled On every N^2 fdm_wait_event() Idle polling set is polled
35 Server performance with different default value of live counter Dispatching throughput and response time (1000 connections, 500 req/s, 1K reply size) Reply rate Response time Reply rate (replys/sec) Response time (ms) Value of live counter (N) Small N Is good Fd in Idle set Fd in doze set Fd in active set Large N useless
Scalability of Linux Event-Dispatch Mechanisms
Scalability of Linux Event-Dispatch Mechanisms Abhish ek Chandra, David Mosberger Internet and Mobile Systems Laboratory HP Laboratories Palo Alto HPL-2-174 December 14 th, 2* E-mail: abh ish ek@cs.umass.edu,
More informationMotivation. Threads. Multithreaded Server Architecture. Thread of execution. Chapter 4
Motivation Threads Chapter 4 Most modern applications are multithreaded Threads run within application Multiple tasks with the application can be implemented by separate Update display Fetch data Spell
More informationException-Less System Calls for Event-Driven Servers
Exception-Less System Calls for Event-Driven Servers Livio Soares and Michael Stumm University of Toronto Talk overview At OSDI'10: exception-less system calls Technique targeted at highly threaded servers
More informationI/O Systems. Amir H. Payberah. Amirkabir University of Technology (Tehran Polytechnic)
I/O Systems Amir H. Payberah amir@sics.se Amirkabir University of Technology (Tehran Polytechnic) Amir H. Payberah (Tehran Polytechnic) I/O Systems 1393/9/15 1 / 57 Motivation Amir H. Payberah (Tehran
More informationAsynchronous Events on Linux
Asynchronous Events on Linux Frederic.Rossi@Ericsson.CA Open System Lab Systems Research June 25, 2002 Ericsson Research Canada Introduction Linux performs well as a general purpose OS but doesn t satisfy
More informationDevice-Functionality Progression
Chapter 12: I/O Systems I/O Hardware I/O Hardware Application I/O Interface Kernel I/O Subsystem Transforming I/O Requests to Hardware Operations Incredible variety of I/O devices Common concepts Port
More informationChapter 12: I/O Systems. I/O Hardware
Chapter 12: I/O Systems I/O Hardware Application I/O Interface Kernel I/O Subsystem Transforming I/O Requests to Hardware Operations I/O Hardware Incredible variety of I/O devices Common concepts Port
More informationChapter 13: I/O Systems
Chapter 13: I/O Systems DM510-14 Chapter 13: I/O Systems I/O Hardware Application I/O Interface Kernel I/O Subsystem Transforming I/O Requests to Hardware Operations STREAMS Performance 13.2 Objectives
More informationChapter 13: I/O Systems. Operating System Concepts 9 th Edition
Chapter 13: I/O Systems Silberschatz, Galvin and Gagne 2013 Chapter 13: I/O Systems Overview I/O Hardware Application I/O Interface Kernel I/O Subsystem Transforming I/O Requests to Hardware Operations
More informationChapter 12: I/O Systems
Chapter 12: I/O Systems Chapter 12: I/O Systems I/O Hardware! Application I/O Interface! Kernel I/O Subsystem! Transforming I/O Requests to Hardware Operations! STREAMS! Performance! Silberschatz, Galvin
More informationChapter 13: I/O Systems
Chapter 13: I/O Systems Chapter 13: I/O Systems I/O Hardware Application I/O Interface Kernel I/O Subsystem Transforming I/O Requests to Hardware Operations STREAMS Performance Silberschatz, Galvin and
More informationChapter 12: I/O Systems. Operating System Concepts Essentials 8 th Edition
Chapter 12: I/O Systems Silberschatz, Galvin and Gagne 2011 Chapter 12: I/O Systems I/O Hardware Application I/O Interface Kernel I/O Subsystem Transforming I/O Requests to Hardware Operations STREAMS
More informationby I.-C. Lin, Dept. CS, NCTU. Textbook: Operating System Concepts 8ed CHAPTER 13: I/O SYSTEMS
by I.-C. Lin, Dept. CS, NCTU. Textbook: Operating System Concepts 8ed CHAPTER 13: I/O SYSTEMS Chapter 13: I/O Systems I/O Hardware Application I/O Interface Kernel I/O Subsystem Transforming I/O Requests
More informationSEDA: An Architecture for Well-Conditioned, Scalable Internet Services
SEDA: An Architecture for Well-Conditioned, Scalable Internet Services Matt Welsh, David Culler, and Eric Brewer Computer Science Division University of California, Berkeley Operating Systems Principles
More informationLecture 8: Other IPC Mechanisms. CSC 469H1F Fall 2006 Angela Demke Brown
Lecture 8: Other IPC Mechanisms CSC 469H1F Fall 2006 Angela Demke Brown Topics Messages through sockets / pipes Receiving notification of activity Generalizing the event notification mechanism Kqueue Semaphores
More informationTopics. Lecture 8: Other IPC Mechanisms. Socket IPC. Unix Communication
Topics Lecture 8: Other IPC Mechanisms CSC 469H1F Fall 2006 Angela Demke Brown Messages through sockets / pipes Receiving notification of activity Generalizing the event notification mechanism Kqueue Semaphores
More informationChapter 8: I/O functions & socket options
Chapter 8: I/O functions & socket options 8.1 Introduction I/O Models In general, there are normally two phases for an input operation: 1) Waiting for the data to arrive on the network. When the packet
More informationChapter 13: I/O Systems
Chapter 13: I/O Systems I/O Hardware Application I/O Interface Kernel I/O Subsystem Transforming I/O Requests to Hardware Operations Streams Performance Objectives Explore the structure of an operating
More informationLecture 15: I/O Devices & Drivers
CS 422/522 Design & Implementation of Operating Systems Lecture 15: I/O Devices & Drivers Zhong Shao Dept. of Computer Science Yale University Acknowledgement: some slides are taken from previous versions
More informationSystems Architecture II
Systems Architecture II Topics Interfacing I/O Devices to Memory, Processor, and Operating System * Memory-mapped IO and Interrupts in SPIM** *This lecture was derived from material in the text (Chapter
More informationThe control of I/O devices is a major concern for OS designers
Lecture Overview I/O devices I/O hardware Interrupts Direct memory access Device dimensions Device drivers Kernel I/O subsystem Operating Systems - June 26, 2001 I/O Device Issues The control of I/O devices
More informationChapter 13: I/O Systems
Chapter 13: I/O Systems Chapter 13: I/O Systems I/O Hardware Application I/O Interface Kernel I/O Subsystem Transforming I/O Requests to Hardware Operations Streams Performance 13.2 Silberschatz, Galvin
More informationChapter 13: I/O Systems
Chapter 13: I/O Systems I/O Hardware Application I/O Interface Kernel I/O Subsystem Transforming I/O Requests to Hardware Operations Streams Performance I/O Hardware Incredible variety of I/O devices Common
More informationChapter 13: I/O Systems
Chapter 13: I/O Systems Chapter 13: I/O Systems I/O Hardware Application I/O Interface Kernel I/O Subsystem Transforming I/O Requests to Hardware Operations Streams Performance 13.2 Silberschatz, Galvin
More informationChapter 13: I/O Systems. Chapter 13: I/O Systems. Objectives. I/O Hardware. A Typical PC Bus Structure. Device I/O Port Locations on PCs (partial)
Chapter 13: I/O Systems Chapter 13: I/O Systems I/O Hardware Application I/O Interface Kernel I/O Subsystem Transforming I/O Requests to Hardware Operations Streams Performance 13.2 Silberschatz, Galvin
More informationModule 11: I/O Systems
Module 11: I/O Systems Reading: Chapter 13 Objectives Explore the structure of the operating system s I/O subsystem. Discuss the principles of I/O hardware and its complexity. Provide details on the performance
More informationComparing and Evaluating epoll, select, and poll Event Mechanisms
Appears in the Proceedings of the Ottawa Linux Symposium, Ottawa, Canada, July, 24 Comparing and Evaluating epoll,, and poll Event Mechanisms Louay Gammo, Tim Brecht, Amol Shukla, and David Pariag University
More informationProceedings of the 4th Annual Linux Showcase & Conference, Atlanta
USENIX Association Proceedings of the 4th Annual Linux Showcase & Conference, Atlanta Atlanta, Georgia, USA October 10 14, 2000 THE ADVANCED COMPUTING SYSTEMS ASSOCIATION 2000 by The USENIX Association
More informationCapriccio : Scalable Threads for Internet Services
Capriccio : Scalable Threads for Internet Services - Ron von Behren &et al - University of California, Berkeley. Presented By: Rajesh Subbiah Background Each incoming request is dispatched to a separate
More informationEmbedded Systems Dr. Santanu Chaudhury Department of Electrical Engineering Indian Institute of Technology, Delhi
Embedded Systems Dr. Santanu Chaudhury Department of Electrical Engineering Indian Institute of Technology, Delhi Lecture - 13 Virtual memory and memory management unit In the last class, we had discussed
More informationFlash: an efficient and portable web server
Flash: an efficient and portable web server High Level Ideas Server performance has several dimensions Lots of different choices on how to express and effect concurrency in a program Paper argues that
More informationCS350: Final Exam Review
University of Waterloo CS350: Final Exam Review Gwynneth Leece, Andrew Song, Rebecca Putinski Winter, 2010 Intro, Threads & Concurrency What are the three views of an operating system? Describe them. Define
More informationIntroduction to Asynchronous Programming Fall 2014
CS168 Computer Networks Fonseca Introduction to Asynchronous Programming Fall 2014 Contents 1 Introduction 1 2 The Models 1 3 The Motivation 3 4 Event-Driven Programming 4 5 select() to the rescue 5 1
More informationOperating System: Chap13 I/O Systems. National Tsing-Hua University 2016, Fall Semester
Operating System: Chap13 I/O Systems National Tsing-Hua University 2016, Fall Semester Outline Overview I/O Hardware I/O Methods Kernel I/O Subsystem Performance Application Interface Operating System
More informationChe-Wei Chang Department of Computer Science and Information Engineering, Chang Gung University
Che-Wei Chang chewei@mail.cgu.edu.tw Department of Computer Science and Information Engineering, Chang Gung University l Chapter 10: File System l Chapter 11: Implementing File-Systems l Chapter 12: Mass-Storage
More informationAnswer to exercises chap 13.2
Answer to exercises chap 13.2 The advantage of supporting memory-mapped I/O to device-control registers is that it eliminates the need for special I/O instructions from the instruction set and therefore
More informationOperating Systems 2010/2011
Operating Systems 2010/2011 Input/Output Systems part 1 (ch13) Shudong Chen 1 Objectives Discuss the principles of I/O hardware and its complexity Explore the structure of an operating system s I/O subsystem
More informationInput / Output. School of Computer Science G51CSA
Input / Output 1 Overview J I/O module is the third key element of a computer system. (others are CPU and Memory) J All computer systems must have efficient means to receive input and deliver output J
More informationComparing the Performance of Web Server Architectures
Comparing the Performance of Web Server Architectures David Pariag, Tim Brecht, Ashif Harji, Peter Buhr, and Amol Shukla David R. Cheriton School of Computer Science University of Waterloo, Waterloo, Ontario,
More informationImproving Scalability of Processor Utilization on Heavily-Loaded Servers with Real-Time Scheduling
Improving Scalability of Processor Utilization on Heavily-Loaded Servers with Real-Time Scheduling Eiji Kawai, Youki Kadobayashi, Suguru Yamaguchi Nara Institute of Science and Technology JAPAN Motivation
More informationCSE 120 Principles of Operating Systems
CSE 120 Principles of Operating Systems Spring 2018 Lecture 15: Multicore Geoffrey M. Voelker Multicore Operating Systems We have generally discussed operating systems concepts independent of the number
More informationModule 12: I/O Systems
Module 12: I/O Systems I/O hardwared Application I/O Interface Kernel I/O Subsystem Transforming I/O Requests to Hardware Operations Performance 12.1 I/O Hardware Incredible variety of I/O devices Common
More informationChapter 13: I/O Systems
COP 4610: Introduction to Operating Systems (Spring 2015) Chapter 13: I/O Systems Zhi Wang Florida State University Content I/O hardware Application I/O interface Kernel I/O subsystem I/O performance Objectives
More informationBackground: Operating Systems
Background: Operating Systems Brad Karp UCL Computer Science CS GZ03 / M030 9 th October 2015 Outline Goals of an operating system Sketch of UNIX User processes, kernel Process-kernel communication Waiting
More informationThe latency of user-to-user, kernel-to-kernel and interrupt-to-interrupt level communication
The latency of user-to-user, kernel-to-kernel and interrupt-to-interrupt level communication John Markus Bjørndalen, Otto J. Anshus, Brian Vinter, Tore Larsen Department of Computer Science University
More informationBackground: I/O Concurrency
Background: I/O Concurrency Brad Karp UCL Computer Science CS GZ03 / M030 5 th October 2011 Outline Worse Is Better and Distributed Systems Problem: Naïve single-process server leaves system resources
More informationG Robert Grimm New York University
G22.3250-001 Receiver Livelock Robert Grimm New York University Altogether Now: The Three Questions What is the problem? What is new or different? What are the contributions and limitations? Motivation
More informationGFS: The Google File System
GFS: The Google File System Brad Karp UCL Computer Science CS GZ03 / M030 24 th October 2014 Motivating Application: Google Crawl the whole web Store it all on one big disk Process users searches on one
More informationDistributed Systems Operation System Support
Hajussüsteemid MTAT.08.009 Distributed Systems Operation System Support slides are adopted from: lecture: Operating System(OS) support (years 2016, 2017) book: Distributed Systems: Concepts and Design,
More informationLast class: Today: Course administration OS definition, some history. Background on Computer Architecture
1 Last class: Course administration OS definition, some history Today: Background on Computer Architecture 2 Canonical System Hardware CPU: Processor to perform computations Memory: Programs and data I/O
More informationInput/Output Systems
Input/Output Systems CSCI 315 Operating Systems Design Department of Computer Science Notice: The slides for this lecture have been largely based on those from an earlier edition of the course text Operating
More informationSilberschatz and Galvin Chapter 12
Silberschatz and Galvin Chapter 12 I/O Systems CPSC 410--Richard Furuta 3/19/99 1 Topic overview I/O Hardware Application I/O Interface Kernel I/O Subsystem Transforming I/O requests to hardware operations
More informationCS370 Operating Systems
CS370 Operating Systems Colorado State University Yashwant K Malaiya Fall 2016 Lecture 2 Slides based on Text by Silberschatz, Galvin, Gagne Various sources 1 1 2 System I/O System I/O (Chap 13) Central
More informationMultiprocessor Systems. Chapter 8, 8.1
Multiprocessor Systems Chapter 8, 8.1 1 Learning Outcomes An understanding of the structure and limits of multiprocessor hardware. An appreciation of approaches to operating system support for multiprocessor
More informationI/O AND DEVICE HANDLING Operating Systems Design Euiseong Seo
I/O AND DEVICE HANDLING 2016 Operating Systems Design Euiseong Seo (euiseong@skku.edu) I/O Hardware Incredible variety of I/O devices Common concepts Port Bus (daisy chain or shared direct access) Controller
More informationCS533 Concepts of Operating Systems. Jonathan Walpole
CS533 Concepts of Operating Systems Jonathan Walpole Improving IPC by Kernel Design & The Performance of Micro- Kernel Based Systems The IPC Dilemma IPC is very import in µ-kernel design - Increases modularity,
More informationModule 12: I/O Systems
Module 12: I/O Systems I/O Hardware Application I/O Interface Kernel I/O Subsystem Transforming I/O Requests to Hardware Operations Performance Operating System Concepts 12.1 Silberschatz and Galvin c
More informationI/O Systems (3): Clocks and Timers. CSE 2431: Introduction to Operating Systems
I/O Systems (3): Clocks and Timers CSE 2431: Introduction to Operating Systems 1 Outline Clock Hardware Clock Software Soft Timers 2 Two Types of Clocks Simple clock: tied to the 110- or 220-volt power
More informationThe Google File System
The Google File System Sanjay Ghemawat, Howard Gobioff, and Shun-Tak Leung SOSP 2003 presented by Kun Suo Outline GFS Background, Concepts and Key words Example of GFS Operations Some optimizations in
More informationI/O Systems. Jinkyu Jeong Computer Systems Laboratory Sungkyunkwan University
I/O Systems Jinkyu Jeong (jinkyu@skku.edu) Computer Systems Laboratory Sungkyunkwan University http://csl.skku.edu Today s Topics Device characteristics Block device vs. Character device Direct I/O vs.
More informationCOMP 3361: Operating Systems 1 Final Exam Winter 2009
COMP 3361: Operating Systems 1 Final Exam Winter 2009 Name: Instructions This is an open book exam. The exam is worth 100 points, and each question indicates how many points it is worth. Read the exam
More informationOperating Systems 2010/2011
Operating Systems 2010/2011 Input/Output Systems part 2 (ch13, ch12) Shudong Chen 1 Recap Discuss the principles of I/O hardware and its complexity Explore the structure of an operating system s I/O subsystem
More informationOperating System Design Issues. I/O Management
I/O Management Chapter 5 Operating System Design Issues Efficiency Most I/O devices slow compared to main memory (and the CPU) Use of multiprogramming allows for some processes to be waiting on I/O while
More informationCSE 120. Translation Lookaside Buffer (TLB) Implemented in Hardware. July 18, Day 5 Memory. Instructor: Neil Rhodes. Software TLB Management
CSE 120 July 18, 2006 Day 5 Memory Instructor: Neil Rhodes Translation Lookaside Buffer (TLB) Implemented in Hardware Cache to map virtual page numbers to page frame Associative memory: HW looks up in
More informationEmbedded Software Programming
Embedded Software Programming Computer Science & Engineering Department Arizona State University Tempe, AZ 85287 Dr. Yann-Hang Lee yhlee@asu.edu (480) 727-7507 Event and Time-Driven Threads taskspawn (name,
More informationBENCHMARKING LIBEVENT AGAINST LIBEV
BENCHMARKING LIBEVENT AGAINST LIBEV Top 2011-01-11, Version 6 This document briefly describes the results of running the libevent benchmark program against both libevent and libev. Libevent Overview Libevent
More informationAuthors : Ruslan Nikolaev Godmar Back Presented in SOSP 13 on Nov 3-6, 2013
VirtuOS: An operating sytem with kernel virtualization Authors : Ruslan Nikolaev Godmar Back Presented in SOSP 13 on Nov 3-6, 2013 Presentation by Bien Aime MUGABARIGIRA Process Isolation and protection
More informationSEDA: An Architecture for Scalable, Well-Conditioned Internet Services
SEDA: An Architecture for Scalable, Well-Conditioned Internet Services Matt Welsh, David Culler, and Eric Brewer UC Berkeley Computer Science Division mdw@cs.berkeley.edu http://www.cs.berkeley.edu/~mdw/proj/seda
More informationResource Containers. A new facility for resource management in server systems. Presented by Uday Ananth. G. Banga, P. Druschel, J. C.
Resource Containers A new facility for resource management in server systems G. Banga, P. Druschel, J. C. Mogul OSDI 1999 Presented by Uday Ananth Lessons in history.. Web servers have become predominantly
More informationThreads. Still Chapter 2 (Based on Silberchatz s text and Nachos Roadmap.) 3/9/2003 B.Ramamurthy 1
Threads Still Chapter 2 (Based on Silberchatz s text and Nachos Roadmap.) 3/9/2003 B.Ramamurthy 1 Single and Multithreaded Processes Thread specific Data (TSD) Code 3/9/2003 B.Ramamurthy 2 User Threads
More information3.1 Introduction. Computers perform operations concurrently
PROCESS CONCEPTS 1 3.1 Introduction Computers perform operations concurrently For example, compiling a program, sending a file to a printer, rendering a Web page, playing music and receiving e-mail Processes
More informationCS 471 Operating Systems. Yue Cheng. George Mason University Fall 2017
CS 471 Operating Systems Yue Cheng George Mason University Fall 2017 Outline o Process concept o Process creation o Process states and scheduling o Preemption and context switch o Inter-process communication
More informationGFS: The Google File System. Dr. Yingwu Zhu
GFS: The Google File System Dr. Yingwu Zhu Motivating Application: Google Crawl the whole web Store it all on one big disk Process users searches on one big CPU More storage, CPU required than one PC can
More informationCS 5523 Operating Systems: Midterm II - reivew Instructor: Dr. Tongping Liu Department Computer Science The University of Texas at San Antonio
CS 5523 Operating Systems: Midterm II - reivew Instructor: Dr. Tongping Liu Department Computer Science The University of Texas at San Antonio Fall 2017 1 Outline Inter-Process Communication (20) Threads
More informationLearning with Purpose
Network Measurement for 100Gbps Links Using Multicore Processors Xiaoban Wu, Dr. Peilong Li, Dr. Yongyi Ran, Prof. Yan Luo Department of Electrical and Computer Engineering University of Massachusetts
More informationChapter 9 Memory Management
Contents 1. Introduction 2. Computer-System Structures 3. Operating-System Structures 4. Processes 5. Threads 6. CPU Scheduling 7. Process Synchronization 8. Deadlocks 9. Memory Management 10. Virtual
More informationAn analysis of Web Server performance with an accelerator
An analysis of Web Server performance with an accelerator S.Jha, F. Sabrina, B. Samuels, and M. Hassan School of Computer Science and Engineering The University of New South Wales Sydney, NSW 2052 Australia
More informationLecture 13 Input/Output (I/O) Systems (chapter 13)
Bilkent University Department of Computer Engineering CS342 Operating Systems Lecture 13 Input/Output (I/O) Systems (chapter 13) Dr. İbrahim Körpeoğlu http://www.cs.bilkent.edu.tr/~korpe 1 References The
More informationLast 2 Classes: Introduction to Operating Systems & C++ tutorial. Today: OS and Computer Architecture
Last 2 Classes: Introduction to Operating Systems & C++ tutorial User apps OS Virtual machine interface hardware physical machine interface An operating system is the interface between the user and the
More informationLight & NOS. Dan Li Tsinghua University
Light & NOS Dan Li Tsinghua University Performance gain The Power of DPDK As claimed: 80 CPU cycles per packet Significant gain compared with Kernel! What we care more How to leverage the performance gain
More informationSDC 2015 Santa Clara
SDC 2015 Santa Clara Volker Lendecke Samba Team / SerNet 2015-09-21 (2 / 17) SerNet Founded 1996 Offices in Göttingen and Berlin Topics: information security and data protection Specialized on Open Source
More informationMemory management. Requirements. Relocation: program loading. Terms. Relocation. Protection. Sharing. Logical organization. Physical organization
Requirements Relocation Memory management ability to change process image position Protection ability to avoid unwanted memory accesses Sharing ability to share memory portions among processes Logical
More informationChapter 13: I/O Systems
Chapter 13: I/O Systems Silberschatz, Galvin and Gagne 2013! Chapter 13: I/O Systems I/O Hardware" Application I/O Interface" Kernel I/O Subsystem" Transforming I/O Requests to Hardware Operations" STREAMS"
More informationCapriccio: Scalable Threads for Internet Services
Capriccio: Scalable Threads for Internet Services Rob von Behren, Jeremy Condit, Feng Zhou, Geroge Necula and Eric Brewer University of California at Berkeley Presenter: Cong Lin Outline Part I Motivation
More informationVirtual Memory Outline
Virtual Memory Outline Background Demand Paging Copy-on-Write Page Replacement Allocation of Frames Thrashing Memory-Mapped Files Allocating Kernel Memory Other Considerations Operating-System Examples
More informationOS-caused Long JVM Pauses - Deep Dive and Solutions
OS-caused Long JVM Pauses - Deep Dive and Solutions Zhenyun Zhuang LinkedIn Corp., Mountain View, California, USA https://www.linkedin.com/in/zhenyun Zhenyun@gmail.com 2016-4-21 Outline q Introduction
More informationSpeeding up Linux TCP/IP with a Fast Packet I/O Framework
Speeding up Linux TCP/IP with a Fast Packet I/O Framework Michio Honda Advanced Technology Group, NetApp michio@netapp.com With acknowledge to Kenichi Yasukata, Douglas Santry and Lars Eggert 1 Motivation
More informationServers: Concurrency and Performance. Jeff Chase Duke University
Servers: Concurrency and Performance Jeff Chase Duke University HTTP Server HTTP Server Creates a socket (socket) Binds to an address Listens to setup accept backlog Can call accept to block waiting for
More informationChapter 13: I/O Systems
Chapter 13: I/O Systems Chapter 13: I/O Systems I/O Hardware Application I/O Interface Kernel I/O Subsystem Transforming I/O Requests to Hardware Operations Streams Performance 13.2 Silberschatz, Galvin
More informationCISC2200 Threads Spring 2015
CISC2200 Threads Spring 2015 Process We learn the concept of process A program in execution A process owns some resources A process executes a program => execution state, PC, We learn that bash creates
More informationComputer System Overview OPERATING SYSTEM TOP-LEVEL COMPONENTS. Simplified view: Operating Systems. Slide 1. Slide /S2. Slide 2.
BASIC ELEMENTS Simplified view: Processor Slide 1 Computer System Overview Operating Systems Slide 3 Main Memory referred to as real memory or primary memory volatile modules 2004/S2 secondary memory devices
More informationQUESTION BANK UNIT I
QUESTION BANK Subject Name: Operating Systems UNIT I 1) Differentiate between tightly coupled systems and loosely coupled systems. 2) Define OS 3) What are the differences between Batch OS and Multiprogramming?
More informationCOSC3330 Computer Architecture Lecture 20. Virtual Memory
COSC3330 Computer Architecture Lecture 20. Virtual Memory Instructor: Weidong Shi (Larry), PhD Computer Science Department University of Houston Virtual Memory Topics Reducing Cache Miss Penalty (#2) Use
More informationOne Server Per City: Using TCP for Very Large SIP Servers. Kumiko Ono Henning Schulzrinne {kumiko,
One Server Per City: Using TCP for Very Large SIP Servers Kumiko Ono Henning Schulzrinne {kumiko, hgs}@cs.columbia.edu Goal Answer the following question: How does using TCP affect the scalability and
More informationWhat s Wrong with the Operating System Interface? Collin Lee and John Ousterhout
What s Wrong with the Operating System Interface? Collin Lee and John Ousterhout Goals for the OS Interface More convenient abstractions than hardware interface Manage shared resources Provide near-hardware
More informationPerformance Impact of Multithreaded Java Server Applications
Performance Impact of Multithreaded Java Server Applications Yue Luo, Lizy K. John Laboratory of Computer Architecture ECE Department University of Texas at Austin 1/2/01 1 Outline Motivation VolanoMark
More informationIsoStack Highly Efficient Network Processing on Dedicated Cores
IsoStack Highly Efficient Network Processing on Dedicated Cores Leah Shalev Eran Borovik, Julian Satran, Muli Ben-Yehuda Outline Motivation IsoStack architecture Prototype TCP/IP over 10GE on a single
More informationProcesses. Process Scheduling, Process Synchronization, and Deadlock will be discussed further in Chapters 5, 6, and 7, respectively.
Processes Process Scheduling, Process Synchronization, and Deadlock will be discussed further in Chapters 5, 6, and 7, respectively. 1. Process Concept 1.1 What is a Process? A process is a program in
More informationAUTOBEST: A United AUTOSAR-OS And ARINC 653 Kernel. Alexander Züpke, Marc Bommert, Daniel Lohmann
AUTOBEST: A United AUTOSAR-OS And ARINC 653 Kernel Alexander Züpke, Marc Bommert, Daniel Lohmann alexander.zuepke@hs-rm.de, marc.bommert@hs-rm.de, lohmann@cs.fau.de Motivation Automotive and Avionic industry
More informationVirtualization, Xen and Denali
Virtualization, Xen and Denali Susmit Shannigrahi November 9, 2011 Susmit Shannigrahi () Virtualization, Xen and Denali November 9, 2011 1 / 70 Introduction Virtualization is the technology to allow two
More information